The Industry-Technology Life Cycle: an Integrating Meta-Model?

THE INDUSTRY-TECHNOLOGY LIFE CYCLE: AN INTEGRATING META-MODEL? Robert U. Ayres International Institute for Applied Systems Analysis Lazenburg, Austria RR-87-3 March 1987 INTERNATIONAL INSTITUTE FOR APPLIED SYSTEMS ANALYSIS Laxenburg, Austria International Standard Book Number 8-7045-0081-x Research Reports, which record research conducted at IIASA, are independently reviewed before publication. However, the views and opinions they express are not necessarily those of the Institute or the National Member Organizations that support it. Copyright @ 1987 International Institute for Applied Systems Analysis All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage or retrieval system, without permission in writing from the publisher. Cover design by Martin Schobel Printed by Novographic, Vienna, Austria Summary This report attempts to answer a rather deep question: To what extent can "pure" economics explain economic growth and technological change? By "pure" economics, it is meant the relationships governing the behavior of abstract enti- ties, producing abstract products or services for sale in an idealized competitive market. Pure economics, in the above sense, admits R&D and innovations of an unspecified kind; it also admits improvements (unspecified) and learning curves or experience curves. The report concludes, however, that the dynamic behavior of the product "life cycle" in specific cases, and the observed clustering of innovations in particular fields at particular times, with periods of rapid progress followed by slow- downs, can only by explained by also taking into account the preexisting state of technology, and the laws of nature. It is argued that technological progress is marked not only by processes of relatively predictable incremental improvement (e.g., "learning curves"), but also by a series of discrete "barriers" and "breakthroughs". These are not random events, although their exact timing is undoubt- edly very difficult to predict. The technological life cycle can be defined as the period from a major breakthrough opening up a new territory for exploitation to the next major bar- rier. It is characterized, in part, by a high initial marginal productivity of R&D, and a more or less continuous decline thereafter, as the territory is gradually exhausted. This model is qualitatively consistent with the well-known "S-shaped curve" phenomenon, describing measures of technological performance over time, as well as a number of other observed phenomena. Foreword Two strands of argument are interwoven in this report. The first strand is that technological innovation is a major driver of economic growth. The second strand is that technological innovation is also a consequence of economic activity. In the jargon of the profession, technological change is essentially endogenous, not exogenous (as has often been assumed for the sake of convenience). In short, technological innovation and economic growth are related like the chicken and the egg: to give either priority over the other is futile. Of course, economists have known this for some time. But more is needed. It is not quite enough to postulate microeconomic mechanisms to explain why entrepreneurs invest in R&D. We also need to understand better the charac- teristic technology life cycle and its relation with the better-known product or industry life cycle. These are major themes of the TES program at IIASA. Pro- fessor Ayres addresses the issue squarely in this report and suggests some direc- tions for future research. THOMAS H. LEE Program Leader, Technology-Economy-Society International Institute for Applied Systems Analysis Preface This paper was begun in late 1984, when the author was in the Department of Engineering and Public Policy at Carnegie-Mellon University in Pittsburgh. It was completed in the winter of 1986, at IIASA, under the auspices of the Technology-Economy-Society Program. Since the subject is central to the TES program, it is being issued as an IIASA Research Report. The author wishes to acknowledge constructive comments on various drafts from Wesley Cohen, Edwin Mansfield, Richard Nelson, Gerhard Rosegger, and Nathan Rosenberg. Thanks are particularly due to Harvey Brooks for his very detailed and helpful critique - almost a minipaper in itself, for which I am most grateful. Any oversimplification, neglect of nuance, or outright error which remains is entirely my own responsibility. ROBERT U. AYRES Deputy Program Leader, Technology-Economy-Society Project Leader, Flexible Manufacturing International Institute for Applied Systems Analysis Contents .. Summary 111 Foreword v Preface vii Background 1 Technological Breakthroughs 7 The Nature of Barriers 11 An Expanding Frontiers Model of the Life Cycle 14 Mechanisms 19 Technological Opportunity 2 1 Economic Implications of the Life Cycle Model 22 Notes 26 References 29 THE INDUSTRY-TECHNOLOGY LIFE CYCLE: AN INTEGRATING META-MODEL? Robert U. Ayres Background The life cycle is a cluster of related phenomena that the economic theory of technological change should explain better. Used in economics, "life cycle" is, of course, a metaphor of succession taken from biology. The stages include concep- tion, birth, infancy, childhood, adolescence, maturity, senescence, and death [I]. Recognizably similar stages can easily be identified in the evolution of other enti- ties, such as a technology [2],a product [3], a firm, an industry, or perhaps even a nation. In fact, without stretching the metaphor unduly, a series of correspon- dences can be identified between the stages of the life cycles for technologies, products, and firms or industries. A stylized description of the life cycle follows. It is summarized in Tables 1-9. The life cycle of an industry begins with a major product innovation or technological breakthrough. Schumpeter saw the innovation as a major entrepreneurial act, preceded by invention and driven by the lure of supernormal monopoly profits. Schumpeter (1912) originally treated the creation of scientific knowledge and invention as exogenous to the economic system [4]. In his later work, Schumpeter (1943) modified this view and allowed for the creation of technology (R&D) as a deliberate activity, especially in large firms. For a recent dis- cussion of the two Schumpeterian models, see Freeman (1982). The innovative product may or may not be immediately successful in the marketplace (infancy); but if it is successful, it quickly spawns both improvements and imitators. The second stage of the cycle (childhood), also described implicitly by Schumpeter, can be characterized as "imitative innovation". At this time there is typically an intense competition among entrepreneurs for market niches, based primarily on design and cost-effectiveness improvements, over the initial entrant. Inventors and innovators try to protect their technologies through patents and secrecy, but diffusion of knowledge occurs inevitably. Only the largest and most dominant firms can expect to capture more than a small proportion of the total benefits of an invention. In this connection, there is a distinct difference between product and process invention. The latter is easier to protect, especially after firms have grown large. However, further innovation during this period of rapid flux is sometimes motivated by a perceived need to invent around a set of dominant patents [5]. Frequently, the competition Table 1. A summary of the life cycle: technology. Logistics, transport, Life Diversity Scale and Labor inventory, cycle versus process require- handling, stage standard technology ments technology INFANCY Unique Custom; High labor Little concern ad hoe intensity, w/ inventory, multi- high-skill1 low relative purpose workers cost of trans- machines needed port, manual handling CHILDHOOD Diversity Small Expansion Inventory costs of types & batch of labor increase sharply, imitators job shop, force, but transport cost manual minimal increase, manual operation "deskilling" handling ADOLESCENCE Increasing Medium to Embodiment High inventory, standardiza- large batch; of labor high transport tion, fewer special skills in costs, semi-auto models; machines & machines handling faster diffusion fixtures MATURITY High degree "Mass" ; Low labor Reduced of standardi- dedicated intensity, inventory, zation, mechani- low skill high approaching zation, for direct transport, saturation transfer mfg. jobs; mechanized lines etc. High skill handling for indirect & managerial jobs SENESCENCE Commodity - like begins even earlier at the R&D stage, as a number of would-be innovators simul- taneously seek to develop a new product for a widely recognized potential market. Entry to the new industry is still easy, in this stage, though many would-be entrants fail. The transition from the second stage (childhood) to the third stage (adolescence) can probably best be characterized as the beginning of consolidation, i.e., the period when the number of entrants to the industry per year is first exceeded by the number of departures. At this period, expansion is most rapid. Eventually, scale economies and the accumulation of "experience" [61 or "learning by doing" achieved by the most successful survivors begins to raise barriers to new entrants. This characterizes the third stage of the cycle, adolescence. The minimum investment required grows larger and the requisite "know- how" resides in a

Load more